European Commission logo
English English
CORDIS - EU research results
CORDIS

Evolution of the social brain: How social complexity affects individual cognition in ants

Periodic Reporting for period 2 - BrainiAnts (Evolution of the social brain: How social complexity affects individual cognition in ants)

Reporting period: 2017-08-01 to 2018-07-31

How does social complexity shape brain structure, its metabolic costs, and the behavioral capability of individuals? The Social Brain Hypothesis posits that larger groups require bigger brains to adaptively process social information, while the Expensive Tissue Hypothesis considers the cost of producing them. Social insects—ants, bees, termites and wasps—live in groups ranging from tens to millions of individuals, have miniaturized brains that support striking individual cognitive abilities and are renowned for their remarkable collective intelligence and division of labor. The distributed processing of information by task-specialized workers might relax the cognitive challenges of individuals, leading to a reduction in brain investment. Additionally, energetic expenses associated with brain size are not well understood in any clade.

Human societies are also characterized by division of labor and collective intelligence. However, it is difficult to examine how social life influenced our brain evolution: we are a single species and historical analyses are limited to comparison of cranial capacity of fossil forms. Ants, in contrast, are a species-rich clade that shows broad variation in social complexity and body size, enabling phylogenetic and socioecological contrasts. Our research can help identifying general principles of the evolutionary neurobiology of social animals, including us. The miniaturized ant brains can inform about evolutionary trade-offs, and how selection for neural efficient and minimal neuroarchitectures can solve cognitive problems and support social behavior. Research on ants has additional societal benefits, for example, by inspiring computer scientists to develop distributed computing systems or inexpensive robots exhibiting swarm intelligence.

We aimed to understand the relationship between brain size, region investment, metabolic costs and individual cognitive abilities of ant workers under different levels of sociality. We examined the effect of distributed information processing, worker polymorphism and colony size on the neurobiological, cognitive and behavioral characteristics of ant workers, and on investment in functionally specialized brain compartments that underlie behavior. To do so, we applied collective animal behavior approaches and tools and neurobiological methods.

We have described anatomical and physiological differences among brains of ant species varying in their social organization and brains of ants from the same colony that perform different social roles. We have also developed technical innovations in brain imaging and brain metabolism to study ant brains. We are designing a learning device that will allow inter and intra species comparisons. This action has strengthened and broadened the research skills and collaboration network of the MSCA fellow, allowing her to find an independent position in Spain.
BrainiAnts was implemented in the laboratory of Prof Traniello (Boston Univ., USA) and the research team of Prof Giurfa (CRCA, France). The MSCA fellow received training in neuroanatomical, neurophysiological, learning and computational imaging techniques, and mentoring in scientific writing, statistical analysis, peer review and publishing, and grant proposal preparation and evaluation. She used complementary approaches to analyze the effect of social complexity on the capabilities of individual ants.
BRAIN STRUCTURE
We collected brain images of workers of different ant species and subcastes. The ant brain integrates functionally specialized regions and association areas in which multimodal processing takes place; we hypothesize sociality might impact differential regional investment among subcastes and between species. Data included:
-Volumetric analysis of sister clades varying in social complexity: Data collection from three pairs of ant species that share similar ecologies but vary in typical group size and colony organization
-Volumetric analysis in polymorphic species to analyze the effect of morphological division of labor: Study of anatomical differences between subcastes in several species to identify general trends independent of ecological contexts
The project also featured significant methodological innovation: To calculate brain subregion volumes, brain images are typically manually labeled; this is very time consuming and sensitive to human bias. We have created, with Dr Arganda-Carreras (Basque country Univ., Spain), statistical brain atlases of ants that allow to have statistical representations of species and also to automatically label samples, reducing time and improving accuracy.
NEUROMODULATION OF SOCIAL BEHAVIOR
We studied the neuromodulation of social behavior of ant species and subcastes by the biogenic amines octopamine, dopamine and serotonin. Model species included ants varying in the type of polymorphism (continuous or discrete). We found aminergic patterns seem to be species specific. We also completed a meta-analysis of the published data of the function of biogenic amines in insects.
BRAIN METABOLIC COST
In collaboration with Dr Waters and Dr Tipping (Providence College, USA), we adapted the Agilent Seahorse respiration system to quantify ex-vivo the brain metabolic activity. We successfully measured metabolic activity from workers varying intra- and interspecifically in body and brain size, and degree of sociality. Our results show that isolated brains can live for hours and that bigger brains measured to date have lower metabolic activity than expected given their mass. Additionally, we used different nutritional regimes (including food deprivation) to examine how ant brain metabolism responds to diet changes.
BEHAVIOR
We designed a robust and simple behavioral assessment to evaluate the effect of division of labor on individual capabilities; we measured differences in activity (average velocity normalized by body length) between worker subcastes of two species of the polymorphic ant genus Pheidole. We found activity differences between subcastes that are consistent between the two species.
We are also developing an aversive learning protocol (based on techniques used for bees) that is suitable for ants of different sizes.
Our research results are disseminated through the publication of peer-reviewed scientific articles in high-quality journals, in national/international conferences/seminars, and through public outreach.
Our results suggest that worker task ecology and division of labor significantly impact brain structure. To separate the role of sociality from ecological influences, we need to analyze more species and other neurobiological variables. We have developed methodological innovations enabling us to do so:
- Brain atlases: Apart from species specific atlases, we can accurately use artificial species-hybrid atlases to automatically trace brain structure in several ant species. We work more rapidly and efficiently, with less human bias, gaining new insights into the conservation and differentiation of brain anatomy patterns in relation to social behavior within and between species
-New method for quantifying ant brain metabolism: We can now study the role of variables such as body and group size, polymorphism, and nutrition in the metabolic cost and regional investment of miniaturized brains
-New learning protocol for ants: Inter and inter species comparisons will be possible through a new aversive conditioning protocol